[rabbitmq-discuss] Fw: high availability solutions?

Fri Jun 17 19:51:40 BST 2011

> We all agree that leaking details of the cluster to the clients is very
> likely not what's wanted in most cases, especially node names. However,
> the numCopies is far too inflexible. There are a number of suggestions
> in this area and they will be implemented in the future. However, there
> are genuine reasons why a client may want to control precisely which
> nodes get mirrors on them, or rather, why Rabbit may not be best placed
> to decide which nodes.

That makes sense. Perhaps something that allows Rabbit to be
"rack-aware" (a la Cassandra) by identifying a
rack/facility/group/unit to a node when it's joined to the cluster,
then mirroring can ensure that the copies are spread across nodes in
different "units".

Just want to say thank you for giving us info on y'all's plans early
and being open to feedback.

-J